%%prrvscci%% %%rssivscci%%
Index
1. Configuration Parameters
2. Summary Statistics
3. Graph Results
4. Glossary of Terms


1. Configuration Parameters
Experiment Name
%%expname%%
Testbed Name
%%testbed%%
Experiment Date
%%date%%
Experiment Time
%%time%%
Protocol
%%interface%%
Channel
%%channel%%
Power Level
%%powerlevel%%
Mote type
%%motetype%%
Number of Nodes
%%numnodes%%
Number of Packets
%%numpackets%%
Inter-Packet Interval (IPI)
%%ipi%%
CSMA
%%csma%%
Transmission Type
%%txtype%%
ACKs
%%ack%%
Max Retries
%%maxretries%%
Noise Sampling
%%noise%%
Number of Noise Samples
%%numnoisesamples%%
Noise IPI
%%noiseipi%%
Rate
%%rate%%
Packet size
%%packetsize%%

2. Summary Statistics

Average node out degree
%%out%%
Average node in degree
%%in%%
Perfect Links
%%perfect%%
Good Links
( 90% < PRR < 100%)
%%good%%
Intermediate Links
( 10% ≤ PRR ≤ 90%)
%%intermediate%%
Poor Links
( 0% < PRR < 10%)
%%bad%%
Dead Links
%%dead%%
High-beta links
(bursty)
%%bursty%%
Low-beta links
(independent)
%%independent%%

3. Graph Results

Connectivity graph
Connectivity graph including PRRs

Cumulative Distribution of PRR:
The Packet Reception Ratio (PRR) distribution with PRR calculated on a per link basis.

%%prrvssnr%%
Asymmetries between link PRRs:
Plots of asymmetries in link PRRs. Asymmetries are typically prevalent in networks. Because of asymmetries, bidirectional link estimates tend to work better for routing protocols. The four plots show how these asymmetries evolved over time. Each graph plots a quarter of the experiment.

First quarter
Second quarter
Third quarter
Fourth quarter

4. Glossary of Terms

PRR: Packet reception ratio (PRR) is calculated as the ratio of packets successfully acknowledged to total transmitted packets over a link. The individual link PRRs are averaged over time. The links are unidirectional, i.e. the PRR of Node A to Node B is distinct from PRR of Node B to Node A.

ARR: The reception ratio of ACK packets. Note that the calculation of ARR is conditioned upon successful reception of the corresponding data packet.

RSSI: Calculates the average and standard deviation of the received signal strength indicator (RSSI) over received packets for each link. RSSI is measured in dBm. Note that RSSI is only calculated based on signal strength of successfully received packets.

Noise distribution: Calculates the distribution of noise floor observed across nodes. Noise samples are taken by reading the RSSI register when no traffic is present. Noise floor at a node is taken as the mode of the noise samples at that node. Noise is measured in dBm

CCI: The Chip Correlation Indicator metric is specific to the CC2420 radio chip in sensor networks. This metric gives a measure of the "goodness" of decoded symbols from the physical layer. It is soft information that indicates the amount of error that was corrected in the physical layer decoding. The average and standard deviation of CCI for each link is calculated.

Link asymmetries: This metric measures the asymmetry in packet reception across two opposite links. As an example, for two nodes A and B, this metric measures the difference in PRR between link A to B and link B to A. High link asymmetries can affect the performance of link estimators which assume links to be symmetric.

ETX asymmetries: Similar to link asymmetries, this metric captures the asymmetries in expected transmissions per packet for two opposite links.

SNR asymmetries: Similar to link asymmetries, this metric captures the asymmetries in signal to noise ratio for two opposite links. The link SNR in dB is calculated as the difference between the mean RSSI of successfully received packets and the mode of noise samples.

Beta: Beta is a measure of temporal variations in a link's packet reception. It measures the "burstiness" of a wireless link and has values between 0 and 1. A high beta link is bursty in nature, i.e., successes and failures for such a link tend to occur in bursts, while a low beta link has successes fairly independent of the previous run of successes or failures.

Rho: Rho is a measure of the correlation of packet receptions for every pair of links (two receivers, one transmitter)