Projects:2017s1-102 HF Radio Automated Link Establishment (ALE) Model
High Frequency (HF) radio communications are able to transmit information such as voice wirelessly over long distances. These distances are achieved by refracting HF waves off layers in the ionosphere. Radio Communications Division of Codan Limited produce a range of HF Radio's which use the 2G-ALE protocol. This project involves tool was made which predicts the call blocking probability for a given system. The tool can also be used to propose a range of channels, for a desired call blocking probability. The ionosphere’s refractive properties vary due to factors such as time of day which in-turn change the channel availablilities of the system. Statistical models were developed to predict a system without ionospheric effects, these were verified on two types of radio, Envoy and Sentry. The model was then adapted to add probabilities of channels being viable. Online tools are then able to predict the channels viability for a given location and time.
High Frequency Radios
HF radios communicate at frequency between 3MHz to 30MHz. There are three main propagation modes for HF EM waves, ground waves, tropospheric waves, and sky waves . Ground waves mainly travel short distances  while being close to the surface of the earth . Multiple factors such as antenna height, polarisation, frequency and terrain affect attenuation of ground waves . Tropospheric waves are waves that propagate though the troposphere, the lowest part of the atmosphere . The moisture and temperature of the troposphere refracts the waves back to the surface . Sky waves propagate by refracting off the ionosphere, allowing it to travel long distances .
Ions, charged particles such as electrons, can travel freely in the ionosphere before colliding or combining with a neutral particle in the ionosphere due to low pressure air . Because the ionised particles are mainly produced by ultraviolet rays from the sun , the ionospheric properties change at different times of the day . Therefore, different set of frequencies must be used for communication between different stations at different times . Due to the complexity of HF propagation, an operator with HF knowledge were often required to operate HF radios .
Automated Link Establishment (ALE)
ALE is a standard used in HF radio to automatically initiate a link from one radio to another  without any assistance from the radio operator . This standard provides a quick and reliable method for a network of stations to contact each other . There are currently two ALE standards, 2G ALE and 3G ALE, that are in use .
Before ALE was standardised, there were adaptive HF radios . Adaptive HF radios were proprietary solutions for automatic linking between HF radios . Because those solutions were proprietary, HF radios from different manufacturer were not compatible . 2G ALE standard was later developed with the cooperation between the United States government and the manufacturers  which enabled interoperability of automated link establishment systems for HF radios from different manufacturers . 3G ALE was later introduced which has higher performance while being interoperable with radios with 2G ALE .
All radios with ALE functionality are assigned a callsign . When the radios are idling (i.e. no active link), the radios will constantly scan though a predefined scan list at two to five channels per second  to listen for its callsign . When a call is initiated, the initiator radio will attempt a three-way handshake with the destination radio . The initiator radio chooses the best channel based on data from LQA, a system that measures and stores link quality between radios for different channels in memory , and verifies that it is not in use . The handshake starts with a preamble signal which contains the callsign of the destination radio . When the destination radio hears its callsign, the scanning process will be stopped. It then respond to the initiator, indicating that the call was successful . A final acknowledgement will be sent by the initiator radio to establish the link .
Multiple papers were written to compare 2G ALE and 3G ALE with modelling and simulation. A paper published by Tait et al.  used a range of modelled scenarios compared to 2G ALE and 3G ALE with fast link setup (FLSU). Tait et al.  found that 3G ALE with FLSU performs better in bidirectional networks especially in situations where the ALE network contains many channels. Where lost time was cumulated from linking by the 2G ALE network. However, if the radio network is large, unidirectional, and has a “hub and spoke” topology, 2G ALE and 3G ALE with FLSU have similar performance . Another paper by Johnson  compared 2G ALE and 3G ALE by simulation. The paper concluded that the simulation demonstrated that 3G ALE has improved network size, SNR requirements, and message throughput over 2G ALE .
Most of the objectives for stage one of the project was achieved. The team was able to develop two models and a software to run the models. However, the team did not manage to fully verify the final model with collected data. The team developed for a network of fixed radios without propagation effects two models, conditional probability and model with simulation. While the software was being written to house model, data generation scripts was also being written to generate data from physical radios to verify the model.
While data for model generation was being generated, some unexpected behaviour of the radios were observed. The behaviour was further investigated, and multiple hypothesis was proposed on the behaviour.
Although the verification of the model was performed, it was not complete because only one data collection run was performed. Although the simulator could estimate the call connection probability quite well, there was a slight discrepancy between the result from the model and the data from data generation script in estimating the probability of all channels being busy. Therefore, the data verification was not completed. This was mainly due to time restrictions.
One of the objective for stage one of the project was not fully achieved; the simulator was not verified. The verification of the simulator needs to be performed before moving to the seconds and the third stage of the project. More test needs to be performed on the simulator for it to be verified.
The next stage of the project is to process with stage two and stage three of the project. This stage of the project is to implement ionospheric propagation effects to the to the model. VOACAP software could be used to predict the ionosphere which can assist in predicting the blocking probability of channels.
- Langford White
- Paul Hirschausen
- Xuyang Jie
- Andrew Thornton
- Chung Seng Foong
- J. S. Seybold, Introduction to RF Propagation, John Wiley & Sons, 2005.
- “Introduction to HF Radio Propagation,” [Online]. [Accessed 1 April 2017].
- D. J. Lusis, “HF Propagation: The Basics,” QST, pp. 11-15, 1983.
- R. E. Menold, “ALE-THE COMING OF AUTOMATIC LINK ESTABLISHMENT,” QST, vol. 79, no. 2, pp. 1-3, 1995.
- “Interoperability and performance standard for medium and high frequency radio systems,” 1 March 1999. [Online]. [Accessed 20 March 2017].
- “Frequency-adaptive communication systems and networks in MF/HF,” 2002. [Online]. Available: . [Accessed 3 April 2017].
- “Automatic Link Establishment,” HF-LINK, [Online]. [Accessed 21 April 2017].
- D. Tait, A. Gillespie and S. Trinder, “MODELLING 2G AND 3G ALE: A QUANTITATIVE COMPARISON,” in 12th IET International Conference on Ionospheric Radio Systems and Techniques (IRST 2012), York, 2012.
- E. E. Johnson, “Simulation results for third-generation HF automatic link establishment,” in IEEE Military Communications. Conference Proceedings, Alantic City, 1999.