EPLRS Models UserGuide
EPLRS Models UserGuide
(EPLRS)
Models User’s Guide
7.0
Contract DASW01 03 D 0008
Disclaimer: As of October 2007, NETWARS was redesignated by the Program Manager Office as the Joint
Communications Simulation System (JCSS). JCSS was selected as the new industry name to better reflect the
inherent joint communication capabilities of the software. Users should be aware that no software updates were
conducted as part of the software name change.
January 7, 2008
Contents
1 Overview .........................................................................................................1
1.1 Intended Use .......................................................................................................................1
1.2 Feature Summary ..............................................................................................................1
1.3 Theory of Operation ..........................................................................................................3
1.3.1 EPLRS Resources ....................................................................................................3
1.3.2 Waveforms...............................................................................................................7
1.3.3 EPLRS Needlines.....................................................................................................7
1.3.4 EPLRS Network Management (ENM) ..................................................................10
1.3.5 EPLRS Radio Set (RS) ..........................................................................................10
1.4 Feature Description .........................................................................................................11
1.4.1 EPLRS ENM..........................................................................................................11
1.4.2 EPLRS RS..............................................................................................................14
1.5 Limitations/Assumptions.................................................................................................16
1.6 Model List .........................................................................................................................16
1.7 Reference Documents ......................................................................................................17
1.8 Additional Information ...................................................................................................18
2 Configuration ...............................................................................................19
2.1 EPLRS_ENM Attributes.................................................................................................19
2.2 EPLRS_RS Attributes .....................................................................................................20
3 Usage .............................................................................................................21
3.1 Discrete Event Simulation Statistics...............................................................................21
3.2 Traffic Usage ....................................................................................................................22
3.3 Supported Equipment Strings ........................................................................................23
3.4 Example Scenarios ...........................................................................................................24
3.5 Troubleshooting ...............................................................................................................31
List of Tables
iii
1 Overview
This document describes key features, attributes, usage, and statistics of the EPLRS
device models shipped as part of the NETWARS models. As part of NETWARS 2007
model development, there are two devices models which, in combination, provide the
EPLRS functionality. These devices are:
1
JCSS Models User’s Guide
divisions.
CSMA Needlines Supported
MSG Needlines Not supported
HDR Duplex Needlines Supported
LDR Duplex Needlines Not Supported
DAP Needlines Not Supported
Waveforms Supported All 16 waveform modes supported
UTO definition Not Supported Common properties in the simulation
environment can be applied to group of
“selected nodes”
Security features Not supported No Key Distribution, Authentication, …
RS Node
RSID Supported
PVC interfaces Supported Arbitrary number of CSMA and HDR
Duplex interfaces
Ethernet Interface Supported
ADDSI Interface Not Supported
MIL-STD-1553 Interface Not Supported
Serial Interface Not Supported
ENM Node
ENM functionality Supported Only ENM configuration feature is
modeled and functions such as monitoring
and key distribution are not modeled
ENM Radio Node Not Supported The ENM is modeled as a standalone
central configuration node. The ENM
communication is not modeled explicitly.
Extra Features
IP Multicast Not Supported
PVC Interface Auto Supported This optional capability does not exist on
Addressing real devices and its availability in
simulation environments greatly
simplifies the configuration of the RS
nodes
Static Routing Supported External IP addresses up to 2 hops away
from the ETHERNET interface of the RS
are reachable
Dynamic Routing Supported Normally, EPLRS radios use static routes
configured on each radio to reach other
nodes and destinations. However, the
NETWARS model provides the optional
capability of activating RIP or OSPF
routing protocols on the PVC interfaces
and remove the need for static route
definitions
3
JCSS Models User’s Guide
EPLRS allows two choices of 2 msec and 4 msec for the timeslot length in an EPLRS
network. The timeslot mode is universal throughout an EPLRS network and remains
fixed once it is chosen. A Frame is a group of 128 consecutive timeslots. The largest time
division in EPLRS is called an epoch which consists of 256 consecutive frames. Figure 2
displays the time resource structure in EPLRS. The encoded data transmission accounts
for some of the time within a timeslot, and the balance of the remaining time is required
for setup, RF propagation delay and processing. Figure 3 shows a detailed view of how
the 128 timeslots (numbered 0 - 127) are arranged in the frame. The timeslots are
numbered such that the data assigned to timeslot 0 is broadcast first, then timeslot 1, then
timeslot 2, etc. Timeslots within a frame can be viewed as blocks of timeslots in
horizontal or vertical groups. Referring to Figure 3, the sixteen vertical groups of
timeslots in a frame are labeled with a Timeslot Index (TSI) number of 0-15 and the eight
horizontal group timeslots are labeled with a Logical Timeslot (LTS) number of 0
through 7. The ENM allocates time resources for users on an LTS-by-LTS basis, instead
of individual timeslots, to allow the planner to work with manageable blocks of time.
With both Time and Frequency resources available as explained above, the planner of an
EPLRS network can allocate different sets of Logical Timeslots (LTS) and channels to
5
JCSS Models User’s Guide
Figure 5. Example of an EPLRS Time and Frequency Resource Allocation Using Only Channel
Zero of a Five Channel System
7
JCSS Models User’s Guide
CSMA needlines provide many hosts the capability to send data to each other, with user
data rates (for the whole needline) from about 150 bps to 971,520 bps. A CSMA needline
allows a large number of endpoints to broadcast data to other members of the needline on
demand. Source transmit opportunities are not reserved for particular endpoints as they
are in all other types of EPLRS needlines. This feature allows hundreds of endpoints to
source data on a single CSMA needline at different times and each source potentially use
all of the available bandwidth if it is not used by the other nodes.
CSMA needlines can operate in 1 hop, 2 hops, 4 hops, and 6 hops modes where the
number of hops in each mode shows how far (in terms of hops) the source and destination
can be located from each other. CSMA uses what is called as the blocking relaying for
multihop communication. For example, In a 4 hops CSMA needline, the source remain
silent for 3 consecutive available timeslots after each transmission and then sends the
next data segment on the fourth available timeslot. In other words, the source doesn’t
start the next transmission until the previous transmission has reached the maximum
number of hops defined by the needline. The selection of relay nodes in a CSMA
needline is performed in a global fashion and specific nodes will be assigned as “Relay”
nodes by a proprietary “Relay Assignment” algorithm. All transmission by RS nodes on
the needline will be relayed only by the relay nodes and if the relay nodes are selected
properly, all other nodes on the needline will receive any transmission by any RS on that
needline. CSMA needlines are defined for one-way communications and the
transmissions are not acknowledged.
The MSG needline provides hosts with a few- to-many communication capability.
Messages are transmitted by a select group of Source RSs (RSs that are allowed to
transmit data from their hosts to other RSs on the needline) and are carried on the MSG
needline, either directly or through relays, to other RSs on the needline. Although there
can be many RSs designated as sources on an MSG needline, only 16 source RS can be
simultaneously active at any one time. Hosts connected to these source RSs on the MSG
needline have the capability to send data, with guaranteed bandwidth without conflict, to
many hosts, with user data rates from about 37.5 bps to 485,760 bps. In an MSG
LDR duplex needlines provide one-to-one communication path between two RSs. Each
transmission is acknowledged by the destination which ensures that the packets are
received by the other end. These needlines provide high reliability balanced data transfer
between two RSs with data rates from about 20 bps to 16192 bps, each way. LDR duplex
needlines are automatically established via the coordination net. The coordination net
uses relay path finding between the two LDR Duplex endpoints, and new relay paths are
re-negotiated throughout the deployment period as required. The time and frequency
resources for all LDR needlines are reserved before the deployment as a pool of
resources. Individual LDR Duplex needlines will then be assigned some of the pool
resources on an as-needed and availability basis.
HDR Duplex needline provides the most reliable one-to-one communication path
between two RSs (and their connected hosts) compared to all other needline types. HDR
duplex needlines provide hosts with an acknowledged communications that ensures host
data is received by the other host. In HDR duplex needlines the resources are reserved a
priori and in both directions thus the bandwidth is always available for use by each of the
two RSs. The data transfer rate between two hosts can range from about 600 bps to
242,880 bps, each way. This type of needline is similar to LDR duplex needlines, except
the user data rate can be much higher. HDR duplex needlines can use the coordination net
to set up and maintain needlines. HDR duplex can also be pre-planned by directing
specific RSs to act as relays for the needline. The path between the two ends of the
needline is in general determined by a “Path Finding” process which involves
transmission of control packets from one end to discover a path to the destination using
nodes with relay functionality and reverse transmission of the packet to the source to
form a bidirectional path. This process will also involve assignment of timeslots and
possibly frequency channels to each node on the path. These needlines use pipeline
relaying similar to MSG needlines. HDR duplex needlines are pre-planned and the
resources will be assigned specifically to each needline.
9
JCSS Models User’s Guide
Dynamically Assigned PVCs (DAP) are the needlines that are established during the
operation on an on-demand basis and are torn down when not needed. Resources for
DAP needlines are allocated before the deployment as a pool of resources and when
needed, some of those resources will be used to establish specific DAP needlines. DAP
needlines are end-to-end duplex needlines and are established either as HDR or as LDR
needlines.
From the view point of IP hosts, an RS functions as an IP router with a number of wired
interfaces as explained above and one radio interface per each needline that the RS is a
member of. Each needline is defined as a separate IP interface called the PVC interfaces
and an IP address is assigned to it. For each needline, the associated needline-specific
protocols define the PHY and MAC layers for the associated IP interface. Each RS by
default has an Ethernet IP interface and a DAP IP interface. Therefore, as an example, an
RS which takes part in 4 needlines will have one Ethernet interface, one DAP interface
and 4 PVC interfaces all defined as IP interfaces. Although all needline-related interfaces
(including DAP) use the same physical interface, their transmissions are separated in time
and frequency based on the initial definition of the needlines and the resources assigned
to them. The routes on each interface are defined as static routes and no dynamic IP
routing protocol is used by the EPLRS RS on its needline IP interfaces.
Figure 7 shows the attributes of the EPLRS_ENM node model. We use the term System
Parameters for all attributes except the Needline Definitions attribute to emphasize the
division-wide nature of those parameters. System parameters define the fundamental
operation mode for the division and determine which frequency channels can be used by
the division or what is the basic time slot length mode (2ms or 4ms) that will be used in
the division. For quick configuration purposes, a default set of values is assigned to those
attributes that can be used if it matches the requirements of the real-world network.
11
JCSS Models User’s Guide
The Channel Set and Active Channels attributes together define which channels will be
used by the needlines of that division. Division ID is the identifier of the division and The
Hop Mode attribute defines whether the frequency hopping is ON or OFF throughout the
division. Time Slot Length takes one of 2ms and 4ms values and defines the basic slot
length of the division. Network Power Level defines the power level used by each radio if
the Radio Power Level is set to “Network Default”. Dynamic Routing is an attribute that
does not exist in real EPLRS systems and is designed to facilitate quick deployment of
EPLRS scenarios in NETWARS. Section 1.4.2.2 explains in more details how this
attribute works. Coord Net Waveform Mode is an attribute that is not used in simulations
and serves as a placeholder for the same attribute in real systems.
While System Parameters attributes define the basic operational parameters of an EPLRS
division, no data can be transmitted unless at least one needline is defined for that
division and some RS nodes have been listed as members of that needline. Figure 8
shows the structure of the Needline Definitions attribute. Each needline is identified by its
unique name. The Type attribute defines what type of needline (CSMA or HDR Duplex)
is defined under that name. Relay Coverage shows the maximum number of hops a
message can travel from source to destination under that needline. The LTS(s) attribute
defines which of the 8 logical time slots (LTS) is used by the needline for data
transmission. However, not all of the slots within those LTSs will necessarily be used by
the needline and that feature is defined by the combination of Circuit Size and Sub LTS
attributes.
The Circuit Size defines the fraction of the LTS (Full, ½ LTS or, ¼ LTS) that is used by
the needline and Sub LTS shows which fraction (0, 1, 2 or, 3 in case of ¼ LTS and 0 or 1
in case of ½ LTS cases) of a full LTS is used by that needline. For example in Figure 9,
four needlines use LTS 0 but each of them only uses a separate ¼ share of that LTS
defines by the Sub LTS attribute.
The Channel attribute shows which channel, from the active channels defines in the
system parameters, is used by the needline. Waveform is the index of the waveform used
by the needline and LCN is the identifying number of the needline that is used by the
model to map IP addresses to needlines. The “Auto Assigned” option for that attribute
can be used to speed up the deployment of the network model. Once the time and
frequency resources of a needline are defined, the user must specify which RS nodes will
be part of that needline and can send or receive information over the needline. The “All
Radios” option automatically makes all RS nodes under that division be members of the
needline. Otherwise, the hierarchical names of all members (excluding the “top.” part of
the name) should be added one by one under the CSMA Members or the HDR Members
compound attribute depending on the type of the needline. Unlike the CSMA needline
where all members have the same role and their names are added under one list, in a
HDR Duplex needline the two endpoints of the needline need to be explicitly specified
under the Endpoint A and Endpoint B sub-attributes of the HDR Members attribute. Also,
any other RS that can potentially serve as a relay node between the two endpoints needs
to be added under the Relays sub-attribute of the same attribute.
The CSMA_DF needline shown in Figure 8 is a CSMA needlines that always exist by
default under the Needline Definitions attribute. This needline is used by all RS nodes to
exchange control messages with each other and thus should not be deleted.
It was explained before that in a CSMA needline each node can be the source, destination
or just a relay node for a message transmission as long as it is a member of that needline.
A message transmitted by a source RS will be broadcasted by any node that receives it
(flooding) until it traverses the maximum number of hops defined for that needline. In the
HDR Duplex needlines, on the other hand, a specific path between the two endpoints of
the needline is established and only the relay nodes in that path forward each message
only to the next hop in that direction of the path. This method offers a much more
efficient use of the channel compared to the flooding method used in CSMA. However,
reliance of the network on a specific path makes the transmission more susceptible to
mobility and other effects that can result in disconnects along the path. To avoid such
situations, all nodes under a HDR Duplex needline periodically exchange control
messages to discover their neighbor RS nodes and monitor connectivity to them. When a
13
JCSS Models User’s Guide
relay RS detects loss of connectivity with its next hop relay, it will notify the ENM node
so that it can find an alternative path from the source to the destination. The topology of
the network is maintained internally as a connectivity graph by the ENM node so that it
can always find the shortest path between the endpoints of HDR Duplex needlines.
1.4.2 EPLRS RS
The EPLRS RS node is the single radio node models which is used in EPLRS networks
in NETWARS. This node model is an IP-capable node and has two main interfaces
namely, Ethernet and EPLRS radio interfaces. The Ethernet interface is a regular Ethernet
interface under the IP layer that is used to connect the EPLRS radio node to external
networks. Operation of each RS node is defined by the needlines it belongs to. Each RS
node can be a member of arbitrary number of needlines and can receive and send traffic
over each needline separately. Figure 10 shows the attributes of the RS node and
highlights the 4 attributes under the EPLRS category.
The Division ID attribute specifies the division and therefore the ENM node which
carries the system parameters for the RS node. Each RS is identified by its unique RSID
which is a positive integer number. The “Auto Assigned” option can be safely used to
automatically assign valid addresses to RS nodes. The Radio Power Level defines the
transmit power of the radio interface. Finally, the Relay Control Boolean attribute
determines whether this node can perform relay functionality for any of the needlines it
belongs to. The RS attribute structure does not show any relation to specific needlines
1.4.2.1 IP Interfaces
Note: The IP interface structure of the EPLRS_RS node model is automatically set by the
system and the user does not need to make any changes to that.
The EPLRS_RS node model has 2 IP interfaces named ETHERNET and EPLRS. The
EPLRS interface has as many subinterfaces as there are needlines which have that RS
node as their member. The subinterfaces are created every time the needlines are defined
or edited and the “OK” button of the attribute window of the ENM node is pressed. This
means that all RS nodes must be present in the scenario before configuring the
needlines on the ENM node.
1.4.2.2 IP Routing
IP routing in real EPLRS systems is done through configuration of static routes on RSs.
In that scheme, for each destination IP address, the IP address of the PVC interface which
should be used to carry the packet to its destination RS node will be specified as the next-
hop address. In the current EPLRS model, exactly the same scheme can be used by
specifying the secondary IP address of the proper needline subinterface as the next-hop
address when configuring the static route for each destination IP address. Static routes are
configured using the standard methodology provided by the IP module in NETWARS.
Although static routing closely represents the way EPLRS routing is configured in real-
world, it may become a time consuming task in simulations with large number of radio
nodes to configure static routes on all RS nodes. The Dynamic Routing utility provided in
the EPRLS model is developed to overcome this problem. When the Dynamic Routing
attribute of the ENM node is set to “Enabled”, it activates the RIP routing protocol on all
needline subinterfaces on all RS nodes under that division. This allows the RS nodes to
exchange routing information and build their dynamic routing tables without need for
15
JCSS Models User’s Guide
static route definitions. Obviously, this utility should be used with care since it can
potentially introduce different routes in the network from what an engineer would
configure using static routes.
1.4.2.3 Waveforms
The EPLRS model in NETWARS supports 15 different waveforms defined for EPLRS
systems by properly setting the payload size to what specified by each waveform.
Moreover, the 90% point of the packet error rate curve for each waveform matches the
90% throughput threshold defined for that waveform in the documents. Assignment of
waveforms to needlines is done in the Needline Definitions attribute of the ENM node.
All RS nodes which are members of a needline will use the characteristics (payload size,
error curve) of that waveform WHEN sending data over that needline.
1.5 Limitations/Assumptions
Due to the proprietary nature of the protocols and some parameters used in real EPLRS,
some features and values have been developed in a generic way based on the partial
information available for them. These approximations include but are not limited to:
Note: The node model named EPLRS which exists under the same palette belongs to an
older version of EPLRS developed in NETWARS and is not compatible with the new
node models and the EPLRS networks defined by them.
17
JCSS Models User’s Guide
Also, for any questions dealing with OPNET specific functionality, consult the standard
OPNET documentation. This documentation can be accessed inside NETWARS by using
the Help > Documentation > ITGuru Documentation menu.
• Deploying the ENM node and all required RS nodes in the scenario from the palette
• Configuring the ENM System Parameters attribute group
• Configuring the ENM Needline Definitions compound attribute
If a node needs to be added later to the scenario, the Needline Definitions attribute of the
ENM node should be opened and “OK”ed even if no changes to the needlines is needed.
Also, if a change is being made to the ENM System Parameters attributes such as the
Channel Set, Active Channels or the Time Slot Length, all needline definitions must be
revisited and their attributes set again. This is because the previous channels may not be
available any more or for example a change in the time slot length may cause the
previous waveforms not to be available any more.
If an RS node is deleted from the network, the needline definitions need to be revisited
only if that node has been specifically mentioned under either the HDR Members or
CSMA Members of one or more needlines.
The first step to configure the ENM is to specify the division it represents and the time
and frequency resources used by that division. This part is performed by configuring the
System Parameters attributes of the ENM node (Figure 7). The 6-Channel option for the
Channel Set provides the largest number of non-overlapping channels and the Active
Channels is better to be set to “All (6)” unless not all channels are expected to be used by
that division (for example when having two divisions i.e., two ENM nodes, and each of
them using 3 channels).
The Needline Definitions attribute is the place where all needlines are defined and it is
also defined which nodes are members of each needline or, equivalently, which needlines
each node supports. Figures 8, 9, and 10 show different examples of needlines defined for
a system. It is reminded again that the CSMA_DF needline which exists by default under
the Needline Definitions attribute, is used for control messaging and should not be
removed (though its resources may be reduced if the number of nodes under that division
is very low).
19
JCSS Models User’s Guide
1. RSID: This is the unique integer number that identifies each RS node. If set to “Auto
Assigned”, the system will pick a valid unique number for that.
2. Radio Power Level: This value is usually set to “Network Default” so that all nodes
use the same power level as set on the division-wide “Network Power Level”
attribute of the ENM node. However, if individual nodes are desired to have different
power levels, it can be set on this attribute.
3. Relay Control: By default, all RS nodes can relay the received messages over the
CSMA needlines they belong to, to other nodes around them. However, this can be
overridden if not all nodes are required to serve as relays by setting the “Relay
Control” attribute of those nodes to “Disabled”. In that case the node will not perform
any relay functionality and simply drops the packets that are not destined for that
node.
To choose these statistics, right-click on a node or right-click in the workspace, and select
“Choose Individual DES Statistics”. Figure 12 shows the list of the available statistics for
the EPLRS nodes. The statistics include the volume of traffic which is sent to and
21
JCSS Models User’s Guide
received from both the higher layer (IP) and lower layer (PHY, Network) by the EPLRS
MAC process. The queue statistics shows the length of the transit and transmit queues.
The transmit queue holds the packets received form the IP layer until they are sent to the
network. The transit queue holds the packets received form the network and need to be
forwarded to other RS nodes.
Statistic Description
Traffic Rcvd from IP (bits/sec) Rate of the traffic received by the EPLRS module form the IP layer
in bits per second
Traffic Rcvd from IP (pkts/sec) Rate of the traffic received by the EPLRS module form the IP layer
in packets per second
Traffic Rcvd from Network (bits/sec) Rate of the traffic received by the EPLRS module form the EPLRS
network in bits per second
Traffic Rcvd from Network Rate of the traffic received by the EPLRS module form the EPLRS
(pkts/sec) network in packets per second
Traffic Sent to IP (bits/sec) Rate of the traffic sent by the EPLRS module to the IP layer in bits
per second
Traffic Sent to IP (pkts/sec) Rate of the traffic sent by the EPLRS module to the IP layer in
packets per second
Traffic Sent to Network (bits/sec) Rate of the traffic sent by the EPLRS module to the EPLRS
network in bits per second (both IP layer and relay traffic)
Traffic Sent to Network (pkts/sec) Rate of the traffic sent by the EPLRS module to the EPLRS
network in packets per second (both IP layer and relay traffic)
Transmit Queue (bytes) Length of the buffer that is used to hold the received IP packets
(from IP layer) until they are sent over the EPLRS network.
Transit Queue (bytes) Length of the buffer that is used to hold the received EPLRS
packets (from other EPLRS nodes) until they are sent again over
the EPLRS network.
The EPLRS_RS nodes receives generic IP over Ethernet traffic from an external device
through its Ethernet interface and sends it to the IP layer so that it can be sent to the
proper needline’s IP sub-interface and reach the destination RS through the members of
that needline. The traffic inside the EPLRS network (between RS nodes) is in the form of
EPLRS packets and is only recognized by the EPLRS interfaces.
The RS node receives an EPLRS packet from its EPLRS interface. After determining
which needline this packet corresponds to, it will be either queued and sent out again on
the EPLRS network (Relay functionality) or, alternatively, the received packet will be
queued as a segment of an IP packet and when the IP packet is fully reconstructed, it will
be sent to the IP layer so that it will be forwarded to the Ethernet interface to reach an
external device.
23
JCSS Models User’s Guide
Although the EPLRS RS can be potentially connected to the end IP node through
arbitrary number of IP hops, it should be noted that the static routing option in EPLRS
only supports addresses up to two hops away from the radio. This means that the above
configurations work fine with both static and dynamic routing options. But any additional
IP nodes between the workstation and the EPLRS RS can only be supported when the
dynamic Routing utility is enabled. Finally, if an EPLRS radio is only used as a relay
node, it can perform by itself and does not need to be connected to any external devices.
Figure 15 shows a scenario with one EPLRS division that contains 7 RS OPFACs. Each
OPFAC in this scenario is a combination of an EPLRS RS node, a router connected to its
Ethernet interface and a workstation node connected to the router (Figure 16). This
typical configuration allows multiple IP devices to be connected to an EPLRS RS node
via the router.
25
JCSS Models User’s Guide
The Default Power Level for the radios is set to “Med-Low” and the system uses the 4ms
time slot length option. The Time Slot Length chosen in this part will determine the
available waveforms for the needlines.
There are 2 CSMA needlines in this division named CSMA_1 and CSMA_2 which are
defined under the Needline Definitions attribute of the ENM node as shown in Figure 18.
Both needlines are set to the 4-Hops “Relay Coverage” mode meaning that each packet
will only be relayed 3 times after leaving the source RS and can reach the destination RS
only if it is within a 4-hops “radius” of the source. The choice of the “Relay Coverage” is
a trade of between the communication range and the data capacity of the network. A
higher coverage number results into longer usage of bandwidth for each packet due to
more relaying transmissions of the same packet. The “LTS(s)”, “Circuit Size” and, “Sub
27
JCSS Models User’s Guide
LTS” attributes for the two needlines show that each needline uses 2 full separate Less.
Since all 3 needlines in this ENM use separate LTS resources. There is no need for
assigning different frequency channels to them and the “Channel” attribute is set to 0 for
all of them meaning that they all use channel 0 for communications. The traffic is set
between node EPLRS_3 and EPLRS_6 and also between EPLRS_4 and EPLRS_5. The
two separate needlines in this scenario are in fact introduced to carry these two traffic
flows separately so that they use different resources and don’t interfere with each other.
In this example scenario, nodes EPLRS_0, EPLRS_1 and EPLRS_2 are assumed to be
located in the middle of the network and will be used as relays for both traffic flows. The
CSMA_1 needline carries the traffic between EPLRS_3 and EPLRS_6 and therefore it
only has these two nodes plus the 3 relay nodes as its members (Figure 19).
Similarly, the CSMA_2 needline only has EPLRS_4 and EPLRS_5 and the three relay
nodes as its members (Figure 20).
The next example scenario (Figure 21) shows how the HDR Duplex needlines are
configured. In this network, a dedicated duplex communication line between RS nodes in
EPLRS_0 and EPLRS_1 OPFACs is required. Also, since EPLRS_0 is a mobile node
Figure 21: EPLRS network with an HDR Duplex needline and node mobility representing path
break-up and re-establishment
29
JCSS Models User’s Guide
In this case, the needline type is set to HDR Duplex and one full LTS (LTS 0) is
dedicated to that. HDR Duplex needlines provide 1-Hop (direct communications), 2-Hop,
4-Hop and, 6-Hop relay coverage options.
The last two options (4 and 6 hops) require 2 frequency channels to be assigned to the
needline and only those options will be provided for the “Channel” attribute when the 4
or 6-hops mode are chosen. The HDR endpoints and the relays are configured under the
“HDR Members” attribute as shown in figure 23.
In this case, all nodes have been assigned as potential relays for this needline so that if the
initial path is broken, a new path can be established by using other nodes as relays. The
initial path between EPLRS_0 and EPLRS_1 is expected to be through EPLRS_2,
EPLRS_3 and, EPLRS_4 nodes. However, EPLRS_0 trajectory suggests that the path
will be broken when EPLRS_0 moves towards Waypoint 1 since it cannot reach any
relay node at that point. When EPLRS_0 moves toward Waypoint 2, it will establish
connectivity with EPLRS_4 at some point and a new path to EPLRS_1 will be
established through EPLRS_4. Again, the new path will be lost as the source moves
towards Waypoint 3 but it should be reestablished when EPLRS_0 approaches Waypoint
4 and established connectivity with EPLRS_6. In that case, the path should go through
EPLRS_5 and EPLRS_5.
A look at the collected statistics for this scenario shows the operation of the network as
predicted by the above argument. Figure 24 shows the relayed traffic at nodes EPLRS_2,
Figure 24: Relayed traffic by the intermediate nodes in the HDR Duplex scenario
3.5 Troubleshooting
The relative complexity of EPLRS in terms of its features and options makes it more
susceptible to configuration errors made by the users. The EPLRS model in NETWARS
provides various forms of checking and debugging tools that make it simple for the user
to troubleshoot a misconfigured network. Below, is a suggested troubleshooting checklist
for locating possible problems in an EPLRS network:
• Simulation Aborts: These types of error result in a simulation abort and typically
happen at time 0 when the different attributes of the EPLRS network (ENM and RS
31
JCSS Models User’s Guide
nodes) are being read and processed by the model. Some of the common causes of
this type of errors are as follows:
o Addition of RS nodes after the ENM node has been configured
A number of active attribute handlers on the ENM automatically perform
necessary configurations on the RS node every time the OK button on the
ENM “Needline Definitions” attribute window is pressed. If a node is
added to the scenario after the initial ENM configuration, the ENM
“Needline Definitions” attribute window needs to be opened and “OK”ed
to configure the newly added RS nodes.
o Duplicate LCN numbers for Needlines
If the “LCN” field of a needline definition is set to “Auto Assigned”, it
will be given a unique LCN number by the active attribute handlers and
that number will be written in the attribute cell. If LCN numbers are
assigned manually, user must make sure they are all unique integer
numbers.
o Incorrect node name in the needline members attribute
Under both the “CSMA Members” and “HDR Members” attributes, the
hierarchical name of the RS nodes must be enters. The hierarchical name
starts from the Nw_Top and goes down the subnet hierarchy until it ends
with the name of the RS node. The “top” subnet string is not part of the
hierarchical name.
• Incorrect end-to-end traffic flow: This type of error is not specific to EPLRS and is a
typical problem in any communication network. The following are the suggested
checkpoints to diagnose the problem:
o Statistics: the data rate statistics can be effectively used to trace the flow of traffic
from the IP layer of the source RS node, to the MAC layers of the relay RS nodes
and finally to the IP layer of the destination RS node.
o ODB traces: for more precise debugging and troubleshooting, a number of traces
have been defined which, when activated, provide useful information about the
feature they represent. These traces are as follows:
eplrs_enm: this trace displays the initial information about the needlines
and RS nodes that is prepared by the ENM node.
eplrs_mac: this trace displays information about the operation of the
EPLRS MAC process. Each RS node has a MAC process which creates as
many CSMA or HDR child processes as necessary for that RS node
eplrs_csma: this trace displays information about the operation of the
CSMA needlines
eplrs_hdr: this trace displays information about the operation of the HDR
Duplex needlines
eplrs_pipeline: this trace displays information about the physical layer
operation of the EPLRS performed by the pipeline stages
eplrs_autoaddr: this trace shows detail information about the IP addresses
assigned to IP interfaces of the RS nodes
33