University of Twente Student Theses

Login

Noise-based Frequency Offset Modulation Simulation Model Design : For the analysis of Transmit-Reference Medium Access Control in Multiple Access Ad-hoc Wireless Sensor Networks

Kriele, M.A. (2018) Noise-based Frequency Offset Modulation Simulation Model Design : For the analysis of Transmit-Reference Medium Access Control in Multiple Access Ad-hoc Wireless Sensor Networks.

[img] PDF
18MB
Abstract:The increasing use of wireless devices has led to more need for energy-efficient communication schemes. Recently, more effort have been put in researching lowpower spread spectrum as an energy-efficient method of communication. transmitted reference (TR) is a low-power spread-spectrum technique which was introduced as a promising communication scheme used in short-range transmissions, such as wireless sensor networks. In TR modulation the transmitter sends the information signal along with the spreading signal shifted by a time or frequency offset, which can be demodulated by the receiver by applying the same offset. This simplifies the receiver architecture significantly. The receiver does thus not need to know the spreading signal used, allowing for any kind of spreading signal; including noise, giving birth to noise-based frequency-offset modulation (N-FOM). N-FOM uses pure noise as information bearer. This is advantageous as it is easy to generate and eliminates the need for complex schemes for flattening the spectrum of the transmitted signal. Therefore, N-FOM allows for multiple-access communication by varying the frequency offset used in the receiver. However, due to the self-correlation module in the receiver, mixing terms of possible other concurrent active nodes increase the noise roughly quadratically; limiting the number of possible concurrent active links as bit error rates worsen. Introducing a medium access control (MAC) protocol to regulate the number of concurrent transmissions could assist in overcoming this barier. Transmitted- Reference MAC (TR-MAC) is a protocol specifically designed to work with N-FOM. The protocol regulates the frequency offsets allocated to transmitting nodes and synchronized with them providing each transmitter a non-overlapping transmission opportunity to send packets. This in order to prevent collisions due to frequency offsets selected twice and to reduce too many concurrent active links. The protocol allows both transmitter-driven and receiver-driven communication. Although proven functional, TR-MAC has only been tested with abstractions of the physical layer and hard limits set on the number of concurrent active links. The creation of a new model is required that is not based on hard limits, but rather implements real physical-layer phenomena of the N-FOM physical layer. This in order to test how the physical layer affects the medium access layer in ways that have previously not been accounted for. Based on theTR-MAC simulation model as a starting point, the N-FOM physical layer has been implemented. Physical layer abstractions and hard limits have been removed. A mathematical expression of the N-FOM layer has been used to model the physical layer for simulation. In order to model a more realistic channel, a Bernoulli random process has been implemented for packet error generation to determine packet error probabilities. Simulations have been performed to verify the physical layer to test its limitations and to see the effects on MAC level in a multipleaccess environment. Results show that physical layer simulation in a single-link environment is according to theory. Simulation results follow the theoretical curve on a 99.9% confidence interval. It can thus be assumed the physical layer is implemented according to theory. Furthermore, the limitation of the physical layer was tested. For this test, nodes were put at an equal distance and made increasingly concurrently active. It became evident the physical limit, based on the parameters set, allowed for a maximum of three concurrent active links as a maximum; thus requiring the need for a MAC protocol. Multiple access simulation of the physical layer in conjunction with the MAC protocol has shown the physical layer has significant impact on the throughput of the system. The self-correlating receiver introduces mixing terms that result in a nearly quadratic increase in noise, resulting in saturation of the throughput when the number of active links increases and eventually a decay due to channel contention. The resulting throughput is less than previously resulted from the TR-MAC measurement results due to these mixing terms. Based on the simulation results it can thus be concluded the physical layer has impact on the MAC layer that has previously been unaccounted for. Additional noise terms introduced by the self-correlation receiver have significant impact on the throughput on the system. However, it is shown that the simulator is functional and can be used further to more extensively test the system as a whole. Currently measurements performed were primarily for functionality analysis purposes, and not for practical implementation as only line-of-sight measurements have been performed. Multipath effects still have to be simulated as they did not fall within the scope of this thesis. Additionally the inclusion of a more extensive clear channel assessment state and/or channel coding could improve the system significantly.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:53 electrotechnology, 54 computer science
Programme:Electrical Engineering MSc (60353)
Link to this item:https://purl.utwente.nl/essays/74472
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page