{"id":3630,"date":"2025-09-21T02:21:38","date_gmt":"2025-09-21T02:21:38","guid":{"rendered":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?p=3630"},"modified":"2025-09-21T02:21:39","modified_gmt":"2025-09-21T02:21:39","slug":"federated-adaptation-personalising-rf-thresholds-without-centralising-raw-data","status":"publish","type":"post","link":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?p=3630","title":{"rendered":"Federated Adaptation: Personalising RF Thresholds without Centralising Raw Data"},"content":{"rendered":"\n<figure class=\"wp-block-embed is-type-wp-embed is-provider-spectrcyde wp-block-embed-spectrcyde\"><div class=\"wp-block-embed__wrapper\">\n<blockquote class=\"wp-embedded-content\" data-secret=\"BvTsFTju5n\"><a href=\"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?page_id=3627\">Federated Adaptation: Personalising RF Thresholds without Centralising Raw Data<\/a><\/blockquote><iframe class=\"wp-embedded-content\" sandbox=\"allow-scripts\" security=\"restricted\" style=\"position: absolute; visibility: hidden;\" title=\"&#8220;Federated Adaptation: Personalising RF Thresholds without Centralising Raw Data&#8221; &#8212; Spectrcyde\" src=\"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?page_id=3627&#038;embed=true#?secret=C8jnCT36rz#?secret=BvTsFTju5n\" data-secret=\"BvTsFTju5n\" width=\"600\" height=\"338\" frameborder=\"0\" marginwidth=\"0\" marginheight=\"0\" scrolling=\"no\"><\/iframe>\n<\/div><\/figure>\n\n\n\n<p>RF-augmented reality (RF-AR) wearables enable<br>detection of threats, casualties and anomalies, but they rely on<br>a set of alert thresholds tailored to mission context and user<br>physiology. Presently, these thresholds are either hard coded<br>or tuned manually, limiting adaptability across individuals and<br>environments. Centralizing raw RF biomarker data to train<br>adaptive models raises privacy and compliance concerns, as raw<br>vitals and location may constitute a search under U.S. law [1], [2].<br>Federated learning offers a solution: devices collaboratively train<br>a shared model while maintaining data locally [3], [4]. We propose<br>Federated Adaptation, a framework that tunes alert thresholds on<br>device using reinforcement signals (e.g., user acknowledgments)<br>and aggregates updates via a federated server. Our contributions<br>are:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>We design a personalised threshold adaptation algorithm<br>that leverages local feedback to adjust detection sensitivity<br>and uses federated averaging to produce a global base<br>model.<\/li>\n\n\n\n<li>We integrate the algorithm with our Glass platform and<br>evaluate on Jetson and Pixel hardware under variable<br>mission conditions, measuring false positive\/negative rates,<br>latency and energy.<\/li>\n\n\n\n<li>We demonstrate that federated adaptation reduces false<br>critical alerts by 32 % compared to static thresholds<br>while preserving battery life and complying with privacy<br>constraints.<\/li>\n\n\n\n<li>We provide a reproducible benchmark harness with JSON<br>metrics, standardised traces and one-command figure generation using OpenBench-AR.<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>RF-augmented reality (RF-AR) wearables enabledetection of threats, casualties and anomalies, but they rely ona set of alert thresholds tailored to mission context and userphysiology. Presently, these thresholds are either hard codedor tuned manually, limiting adaptability across individuals andenvironments. Centralizing raw RF biomarker data to trainadaptive models raises privacy and compliance concerns, as rawvitals and location&hellip;&nbsp;<a href=\"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?p=3630\" rel=\"bookmark\"><span class=\"screen-reader-text\">Federated Adaptation: Personalising RF Thresholds without Centralising Raw Data<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":2336,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"neve_meta_sidebar":"","neve_meta_container":"","neve_meta_enable_content_width":"","neve_meta_content_width":0,"neve_meta_title_alignment":"","neve_meta_author_avatar":"","neve_post_elements_order":"","neve_meta_disable_header":"","neve_meta_disable_footer":"","neve_meta_disable_title":"","footnotes":""},"categories":[10],"tags":[],"class_list":["post-3630","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-signal_scythe"],"_links":{"self":[{"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/posts\/3630","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=3630"}],"version-history":[{"count":1,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/posts\/3630\/revisions"}],"predecessor-version":[{"id":3631,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/posts\/3630\/revisions\/3631"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/media\/2336"}],"wp:attachment":[{"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3630"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3630"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3630"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}