AEGAEUM JOURNAL ISSN NO: 0776-3808

Convert Radio-waves to alert using SDR, AWS Lambda and Amazon Transcribe API

Ark Vrajlal Balar & Dr. Kamaljit I. Lakhtaria

Department of Computer Science, Rollwala Computer Center, Gujarat University

Abstract- Communication is essential and the major aspect Conversations and commands conveyed over the radio nowadays. The most used and authentic communication method link could be transcripted, it would be an add-on is radio communication. Conversations and commands conveyed advantage for its users. over the radio link can be transcripted, so that the receiver can 2. Linux, is a family of open source Unix-like operating have a log of the conversation. From the generated transcript systems based on the Linux kernel, an conversations are read again to find if any significant words or kernel first released on September 17, 1991, by Linus commands are conveyed. If specific commands are conveyed; it Torvalds. Linux is typically packaged in a Linux can be used for creating an alert or for logging. distribution.A common feature of Unix-like systems, Linux includes traditional specific-purpose Index Terms- Radio, radio communication, sdr, aws, programming languages targeted at scripting, text transcription, reading, alert. processing and system configuration and management in general. Linux distributions support shell script. A scripting or script language is a programming I. INTRODUCTION language for a special run-time environment that Communication is the important aspect nowadays. automates the execution of tasks; the tasks could Communication takes place in everything, everywhere. The most alternatively be executed one-by-one by a human used and authentic communication method is radio operator. Primitives are usually the elementary tasks or communication. Radio communication is mostly used in defence API calls, and the language allows them to be combined and plays a game changing role in it. into more programs. Environments that can be Close communication between the data analysts and data automated through scripting include software producers will be a necessary and integral nexus if optimal applications, web pages within a web browser, usage of clinical benefit is to be achieved. the shells of operating systems (OS), embedded 1. Amateur radio, also known as ham radio, is the use of systems, as well as numerous games. A scripting radio frequency spectrum for purposes of exchange of language can be viewed as a domain-specific language messages, wireless experimentation, self-training, for a particular environment; in the case of scripting an private recreation, radio sport, contesting, and application, it is also known as an extension language. emergency communication. Scripting languages are also sometimes referred to as SDR - Software Defined Radio, solutions are beneficial very high-level programming languages, as they operate in the military sector, where communication systems at a high level of abstraction, or as control languages, need to have a longer service life time than in the particularly for job control languages on mainframes. commercial sector, SDR helps to protect investments by 3. Amazon Web Services (AWS) is a subsidiary of prolonging the useful service life of communication Amazon that provides on-demand cloud computing systems. platforms and APIs to individuals, companies, and A major motivation within the commercial governments, on a metered pay-as-you-go basis. In communications arena is the rapid evolvement of aggregate, these cloud computing web services provide communications standards, making software upgrades a set of primitive abstract technical infrastructure and of base stations a more attractive solution than the distributed computing building blocks and tools. It costly replacement of base stations. Common for both comprises of many services, few of them are: the military and the commercial sector, is that SDR opens up a range of possibilities by making existing A. Amazon S3 types of radio applications easier to implement, and by B. AWS Lambda allowing new types of applications. C. Amazon Transcribe D. Amazon SNS E. Amazon CloudWatch

Volume 8, Issue 7, 2020 http://aegaeum.com/ Page No: 546 AEGAEUM JOURNAL ISSN NO: 0776-3808

A. Amazon S3 or Amazon Simple Storage Service is a or sending of messages to subscribing endpoints or service offered by Amazon Web Services (AWS) that clients. In Amazon SNS, there are two types of provides object storage through a web service interface. clients—publishers and subscribers—also referred to as B. AWS Lambda is an event-driven, serverless computing producers and consumers. platform provided by Amazon as a part of Amazon Web E. Amazon CloudWatch monitors your Amazon Web Services. It is a computing service that runs code in Services (AWS) resources and the applications you run response to events and automatically manages the on AWS in real time. You can use CloudWatch to computing resources required by that code. collect and track metrics, which are variables you can C. Amazon Transcribe uses advanced machine learning measure for your resources and applications. technologies to recognize speech in audio files and This study is a part of an attempt to upgrade the communication transcribe them into text. You can use Amazon systems integrating Linux and Amazon Web Services(AWS). Transcribe to convert audio to text and to create applications that incorporate the content of audio files. D. Amazon Simple Notification Service (Amazon SNS) is a web service that coordinates and manages the delivery

II. RESEARCH FLOW A radio station is usually equipped with all radio communication equipment for good reception. SDR is a good option for reception and for recording purposes. SDR options available for the purpose are as follows: 1. SDR# 2. HDSDR 3. SDR-RADIO.COM V2/V3 4. Linrad 5. GQRX 6. CubicSDR

7. Studio1 Image-1: SDR scanner tool - GQRX 8. SDRUno 9. SigDigger A. SDR scanner tool - GQRX 10. ShinySDR GQRX is an open source software defined radio receiver (SDR) 11. WebRadio powered by the GNU radio and the graphical toolkit. GQRX 12. OpenWebRX supports many of the SDR hardware available, including Airspy, 13. Sodira Funcube Dongles, rtl-sdr, HackRF and USRP devices. Here, 14. SDR Touch RTL-SDR 2832U DVB-T(Digital Video Broadcasting — 15. Wavesink Plus Terrestrial) USB dongle with monopole antenna is used for the 16. RFAnalyzer reception of radio signals. 17. cuSDR 18. PowerSDR 19. QtRadio 20. Multimode 21. Sdrangelove 22. kukuruku 23. ham2mon Here, the SDR scanner tool used is GQRX.

Volume 8, Issue 7, 2020 http://aegaeum.com/ Page No: 547 AEGAEUM JOURNAL ISSN NO: 0776-3808

C. AWS Amazon Web Services is a subsidiary of Amazon that provides on-demand cloud computing platforms and APIs. AWS services and APIs used in this work are as follows: 1. Amazon S3 2. AWS Lambda 3. Amazon Transcribe API 4. Amazon SNS 5. Amazon CloudWatch

1. Amazon S3 stores data as objects within buckets. An object is a file and any optional metadata that describes the file. To store a file in Amazon S3, one uploads it to a bucket. 2. AWS Lambda can be used to run the code in response to events, such as changes to data in an Amazon S3 bucket and to run the code in response to HTTP requests using Amazon API Gateway. 3. Amazon Transcribe uses advanced machine learning technologies to recognize speech in audio files and transcribe them into text. You can use Amazon Transcribe to convert audio to text. The output from the

Image-2: RTL-SDR 2832U DVB-T with monopole antenna transcription job is also stored in an S3 bucket. However, radio communication may not be detected in remote Users can create a custom vocabulary to help Amazon areas, the solution for this problem is additional patching to Transcribe recognize words that are specific to your use antennas or different antennas. case and improve its accuracy in converting speech to Features of this SDR tool are as follows: text. ● Discover devices attached to the computer. ● Process I/Q data from the supported devices. ● Change frequency, gain and apply various corrections (frequency, I/Q balance). ● AM, SSB, CW, FM-N and FM-W (mono and 4. Amazon Simple Notification Service (Amazon SNS) is stereo) demodulators. a web service that coordinates and manages the delivery ● Special FM mode for NOAA APT. or sending of messages to subscribing endpoints or ● Variable band pass filter. clients. ● AGC, squelch and noise blankers. In Amazon SNS, there are two types of clients— ● FFT plot and waterfall. publishers and subscribers—also referred to as ● Record and playback audio to / from WAV producers and consumers. Publishers communicate file. asynchronously with subscribers by producing and ● Record and playback raw baseband data. sending a message to a topic, which is a logical access ● Spectrum analyzer mode where all signal point and communication channel.Subscribers consume processing is disabled. or receive the message or notification over one of the ● Basic remote control through TCP connection. supported protocols (that is, Amazon SQS, HTTP/S, ● Streaming audio output over UDP. email, SMS, Lambda) when they are subscribed to the topic. B. Linux - shell scripting A shell script is a computer program designed to be run by the Unix/Linux shell. A shell is a command-line interpreter and typical operations performed by shell scripts include file manipulation, program execution, and printing text. Here, the shell scripting is used for the purposes as follows: 5. Amazon CloudWatch monitors Amazon Web Services 1. Record conversation from SDR tool (AWS) resources and the applications running on AWS 2. Upload the recording to AWS S3 bucket in real time. Users can use CloudWatch to collect and track metrics, which are variables and can measure the resources and applications. As audio recording is successfully uploaded to the Amazon S3 bucket, a lambda function configured on Amazon S3 bucket to trigger at PUT_OBJECT event will be triggered and call the

Volume 8, Issue 7, 2020 http://aegaeum.com/ Page No: 548 AEGAEUM JOURNAL ISSN NO: 0776-3808

[4] Herschel H. Loomis, Jr. George W. Dinolt, and Frank E. Kragh, Amazon Transcribe API which will initiate the transcription “Performance of high-reliability space qualified processors implementing process. As the transcription job is finished, the transcript is put Software Defined Radios” in Naval Postgraduate School, Monterey, California. back to another AWS S3 bucket. [5] Shahrokh Valaee, Chen Feng, Anthea Waln, Sy Au, “System method and Another AWS Lambda function configured on the Amazon S3 computer program for anonymous localization”, US Patent - US9763035B2. bucket containing transcripts at PUT_OBJECT event will be [6] Giuseppe Baruffa, Mauro Feminella, Matteo Pergolesi, Gianluca Reali, “A triggered and read the transcript for specific words. If it will Big Data architecture for spectrum monitoring in cognitive radio,” in Annals of Telecommunications volume 73, pages 451–461(2018). detect them it will trigger the Amazon SNS topic to publish an [7] Mamiko Inamori, Anas M. Bostaman, Yukitoshi Sanada, “Influence of alert to its subscribers. sampling jitter on discrete time receivers” in 2005 IEEE 16th International Symposium on Personal, Indoor and Mobile Radio Communication. [8] Tore Ulversoy, “Software Defined Radio: Challenges and Opportunities” in IEEE Communications Surveys & Tutorials (Volume: 12,Issue: 4, Fourth Quarter 2010). [9] Hüseyin Arslan, Joseph Mitola III, “Cognitive radio, software-defined radio,and adaptive wireless systems” in Volume 7, Issue 9, Special Issue: Cognitive Radio, Software Defined Radio and Adaptive Wireless Systems, November 2007. [10] Aditya Gudipati, Daniel Perry, Li Erran Li, Sachin Katti, “SoftRAN: Software defined Radio Access Network” in HotSDN‘13: Proceedings of the second ACM SIGCOMM workshop on Hot topics in software defined networking August 2013 Pages 25–30. [11] Gianmarco Baldini, Igor Nai Fovino, Marcelo Masera, Marco luse, Vincenzo Pellegrini, Enzo Bagagli, Giuseppe Rubino, Raffaele Malangone, Marcoccio, Stefano, Fabio Senesi, “An early warning system for detecting GSM- wireless infrastructure in the high-speed railway infrastructure” in International Journal of Critical Infrastructure Protection, Volume 3, Issue 3-4, December 2010, Pages 140-156. [12] V Keller, “Adaptation and application of a transcript analysis tool to analyze a computer-mediated communication (CMC)” in Athabasca University Library. [13] Ken O Burtch, “Linux Shell Scripting with ” in January 2004 published by Pearson Higher Education. [14] Hobin Yoon, Ada Gavrilovska, Karsten Schwan, Jim Donahue, “Interactive III. CONCLUSION use of cloud services Amazon SQS and S3” in 2012, 12th IEEE/ACM This work has managed to demonstrate the transcription of the International Symposium on Cluster, Cloud and Grid Computing (CCGRID-2012). radio communication with use of SDR tool as radio signal [15] Wei-Tsung Lin, Chandra Krintz, Rich Wolski, Michael Zhang, Xiaogang receiver and Amazon Transcribe as a transcriptor. For the first Cai, Tongjun Li and Wejin Xu, “Tracking causal order in AWS Lambda prototype development, an SDR-tool is used to receive and for applications” in 2018 IEEE International Conference on Cloud Engineering recording. The recordings can easily be uploaded to AWS S3 (IC2E). [16] Marian Kiran, Peter Murphy, Inder Monga, Jon Dugan, Sartaj Singh bucket and transcribed to obtain transcript and read it with AWS Baveja, “Lambda architecture for cost-effective batch and speed Big Data Lambda. processing” in 2015 IEEE International Conference on Big Data. As a future work, to allow at the signal receiving end, use [17] Simson L. Garfinkel, “An Evaluation of Amazon’s Grid Computing Services: EC2, S3 & SQS” in Harvard Computer Science Group Technical different types of antenna and analyze the received signal quality Report, TR - 08-07. and at the transcription end, optimize the transcription time by [18] Deepthi Karkada, Vikram A. Salatore, “Training speech recognition models changing to open source transcription engine and enhance the on HPC Infrastructure” in 2018 IEEE/ACM Machine Learning in HPC Environments (MLHPC). transcription process by enabling speaker identification and [19] Clifford B. Anderson, “On evolving the operations of the Vanderbilt identify different channels. Television News Archive(VTNA): Cloud Computing for AudioVisual Preservation” in Jean and Alexander Heard Library, Vanderbilt University, TN, US. ACKNOWLEDGMENT

I WOULD LIKE TO THANK DR. KAMALJIT LAKHTARIA, DR. HARDIK AUTHORS JOSHI, ADHIR SAIYADH, AND JANAK THAKKAR FOR THEIR HELPFUL COMMENTS. First Author – Ark Vrajlal Balar, MTech (Networking & Communications), REFERENCES Department of Computer Science, Rollwala Computer Center, [1] Lance A. Liotta, Mark Lowenthal, Arpita Mehta, Thomas P. Conrads, Timothy D. Veenstra, David A. Fishman, Emanuel F. Petricoin, III, Gujarat University. “Importance of Communication Between Producers and Consumers of E-mail: [email protected]. Publicly Available Experimental Data” in 16/02/2005, JNCI: Journal of the Second Author – Dr. Kamaljit I. Lakhtaria, National Cancer Institute, Volume 97, Issue 4, 16 February 2005, Pages Associate Professor, 310–314. [2] Mary C. Politi and Richard L. Street, “The importance of communication in Department of Computer Science, collaborative decision making: facilitating shared mind and the Gujarat University. management of uncertainty” in Volume 17, Issue 4, Special Issue: Evidence E-mail: [email protected]. Based Medicine. [3] Ning Zhang, Wenhai Sun, Wenjing Lou, Y. Thomas Hou, Wade Trappe, “ROSTER : Radio context attestation in cognitive radio network” in the Correspondence Author – Ark Vrajlal Balar, 2018 IEEE Conference on Communications and Network Security (CNS).

Volume 8, Issue 7, 2020 http://aegaeum.com/ Page No: 549 AEGAEUM JOURNAL ISSN NO: 0776-3808

MTech (Networking & Communications), Department of Computer Science, Rollwala Computer Center, Gujarat University. E-mail: [email protected], Contact number: (+91) 88 666 606 75

Volume 8, Issue 7, 2020 http://aegaeum.com/ Page No: 550