15-page paper due in 48 hours, where do I start?

You need to get a skeleton on the page right now, then fill it in with the most relevant sources you can locate in the next few hours; that way you won’t waste precious time trying to write a perfect draft before you even know what you’ll be saying. Start by writing a one‑sentence thesis, sketch a three‑section outline that mirrors the typical introduction‑methods‑results‑discussion flow, and then spend the next 30 minutes pulling the five or six most-cited IEEE papers that directly address the core concept of your analysis. Those papers will become the backbone of your literature review and will give you enough citations to meet the length requirement without having to read every article in the pile. The biggest misconception is that you have to read every paper in depth before you can decide whether it belongs in your work. In a time crunch, skim the abstract, conclusion, and any figures that look like they answer your research question; if they do, download the PDF and flag it for a quick read later. Most engineering papers follow a predictable structure, so you can often judge relevance from the first two pages. Trade‑offs appear when you try to be overly comprehensive—adding marginally related sources can bloat your paper and eat up writing time, while being too narrow may leave reviewers questioning the depth of your review. Aim for a balance: a solid core of three to four primary sources plus a handful of secondary references that support or contrast your findings. Edge cases matter, too. If your topic is very niche, you might only find a couple of directly relevant IEEE articles; in that scenario you can broaden the search to conference proceedings or related standards, and you’ll need to be explicit in your paper about why the literature is limited. Conversely, if the field is saturated, you’ll have to be selective and perhaps focus on the most recent or most highly cited works to demonstrate you’re aware of current developments. Your exact situation—how familiar you already are with the subject, how much data you already have, and whether you need to run any simulations—will dictate how much of the 15 pages you allocate to background versus analysis. If you already have experimental results, you can devote more space to interpreting them; if not, the bulk of the paper will be a critical synthesis of the literature you gathered. Imagine you’re writing about “energy‑efficient routing in wireless sensor networks.” You could start by pulling the two seminal IEEE papers that introduced the main algorithmic ideas, then add the most recent survey that compares those ideas to newer machine‑learning approaches. With those three sources you can draft an introduction that frames the problem, a literature section that contrasts the classic and modern methods, and a discussion that points out gaps you’ll explore in your own brief simulation. By the end of the first hour you’ll have a clear roadmap and a handful of citations, and the remaining 47 hours can be spent fleshing out each section rather than hunting for papers. Stick to this focused, step‑by‑step approach and you’ll finish a coherent, well‑referenced 15‑page paper before the deadline.